Bayesian Inference for Unidirectional Misclassification in Ordinal Covariates

• Activity Number: 171
• Type: Contributed
• Date/Time: Monday, August 1, 2016 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Liangrui Sun* and Chaoxiong Xia and Yuanyuan Tang and Shun Takai
• Companies: and Northern Illinois University and Saint Luke's Health System and Northern Illinois University
• Keywords: Bayesian regression models ; partial identification ; Markov chain Monte Carlo ; Unidirectional misclassification

Abstract

In this talk, we study the identification of Bayesian regression models, when an ordinal covariate is subject to unidirectional misclassification. Xia and Gustafson (2016) (Bayesian regression models adjusting for unidirectional covariate misclassification, under review) obtained model identifiability for non-binary regression models, when there is a binary covariate subject to unidirectional misclassification. In the current project, we establish the identifiability of regression models for misclassified ordinal covariates with more than two categories, based on forms of observable moments. Simulation studies are conducted that confirms the theoretical results. A case study is performed using a survey dataset on vehicle demand.